forecasting the incidence and prevalence of patients with...

Research ArticleForecasting the Incidence and Prevalence of Patients withEnd-Stage Renal Disease in Malaysia up to the Year 2040

Mohamad Adam Bujang,1 Tassha Hilda Adnan,1 Nadiah Hanis Hashim,1

Kirubashni Mohan,1 Ang Kim Liong,2 Ghazali Ahmad,3 Goh Bak Leong,2,4

Sunita Bavanandan,3 and Jamaiyah Haniff5

1National Clinical Research Centre, Kuala Lumpur, Malaysia2Clinical Research Centre, Serdang Hospital, Kajang, Malaysia3Department of Nephrology, Kuala Lumpur Hospital, Kuala Lumpur, Malaysia4Department of Nephrology, Serdang Hospital, Kajang, Malaysia5Malaysian Health Performance Unit, Ministry of Health, Kuala Lumpur, Malaysia

Correspondence should be addressed to Mohamad Adam Bujang; [email protected]

Received 6 June 2016; Accepted 22 December 2016; Published 28 February 2017

Academic Editor: Tej Mattoo

Copyright © 2017 Mohamad Adam Bujang et al.This is an open access article distributed under theCreative CommonsAttributionLicense, which permits unrestricted use, distribution, and reproduction in anymedium, provided the originalwork is properly cited.

Background. The incidence of patients with end-stage renal disease (ESRD) requiring dialysis has been growing rapidly inMalaysiafrom 18 per million population (pmp) in 1993 to 231 pmp in 2013. Objective. To forecast the incidence and prevalence of ESRDpatients who will require dialysis treatment in Malaysia until 2040.Methodology. Univariate forecasting models using the numberof new and current dialysis patients, by theMalaysianDialysis andTransplant Registry from 1993 to 2013were used. Four forecastingmodels were evaluated, and themodel with the smallest error was selected for the prediction. Result. ARIMA (0, 2, 1) modeling withthe lowest error was selected to predict both the incidence (RMSE = 135.50, MAPE = 2.85, and MAE = 87.71) and the prevalence(RMSE = 158.79, MAPE = 1.29, andMAE = 117.21) of dialysis patients.The estimated incidences of new dialysis patients in 2020 and2040 are 10,208 and 19,418 cases, respectively, while the estimated prevalence is 51,269 and 106,249 cases. Conclusion. The growth ofESRD patients on dialysis in Malaysia can be expected to continue at an alarming rate. Effective steps to address and curb furtherincrease in new patients requiring dialysis are urgently needed, in order to mitigate the expected financial and health catastrophesassociated with the projected increase of such patients.

1. Introduction

End-stage renal disease (ESRD), now termed chronic kidneydisease (CKD) stage five is a state of permanent loss of renalfunction when measured or calculated glomerular filtrationrate is less than 15ml/min permanently. Worldwide, thenumber of ESRD patients is growing rapidly in developedand developing countries, fueled by aging populations anda pandemic of chronic noncommunicable diseases especiallydiabetesmellitus and hypertension. Current projections indi-cate that, by 2030, the global population of ESRD patientsliving on dialysis may exceed 2 million [1].

InMalaysia, the incidence and prevalence of patients withESRD has been on an upward trend for the past 20 years.The increase in the dialysis population is attributable to the

increasing availability of haemodialysis treatment facilitiesand easier access to public or subsidised funding, especiallyin the nongovernmental sector. The 22nd Report of theMalaysian Dialysis and Transplant Registry in 2013 recordeda total of 32,026 patients receiving dialysis treatment, 29,192on haemodialysis (91%) and 2,834 on peritoneal dialysis (9%)[2]. In 1993, 358 newESRDpatients were treatedwith dialysis,2,629 in 2003, while in 2013 the number steeply increased to6,985. The incidence of patients with ESRD on dialysis was18 per million population (pmp) in 1993. Subsequently, thenumbers doubled from 104 pmp in 2003 to 231 pmp in 2013.The prevalence of patients with ESRD on dialysis was 71 pmpin 1993, rising subsequently more than twofold from 415 pmpin 2003 to 1,059 pmp in 2013 [2].

HindawiInternational Journal of NephrologyVolume 2017, Article ID 2735296, 5 pageshttps://doi.org/10.1155/2017/2735296

https://doi.org/10.1155/2017/2735296

2 International Journal of Nephrology

As the risk of ESRD increases with age, the aging popu-lation in Malaysia is expected to have a large impact on thenumber of incident dialysis patients. A study by Wakasugiet al. observed the trend in the incidence rates of dialysis inJapanese population between 2008 and 2012. The total num-ber of incident dialysis patients was predicted to increase by12.8% from 36,590 in 2012 to 41,270 in 2025, with higherincrement in the oldest age group (≥85 years). Male andfemale patients were expected to increase by 92.6% and62.2%, respectively [3].

Looking at the trend in the past 10 years, there is a need todo forecasting studies of ESRD patients to create awarenessamong the policy makers and the public with regard to themagnitude of the disease. Accurate prediction can be usedto facilitate the policy making, decisions on optimal healthcare allocation and appropriate financial planning and allowfor more appropriate use of health care resources for theMalaysian population. This study aims to use time-seriestechniques on the available historical data to find the bestmodel with the smallest error that can forecast the numberof new ESRD patients in Malaysia in the future. Hence, thepurpose of this study is to describe the trend of incidence andprevalence of ESRD patients in Malaysia from 1993 to 2013.The study also aims to determine the best univariate forecast-ing model which produces the smallest root mean squarederror (RMSE), mean absolute percentage error (MAPE), andmean absolute error (MAE), subsequently to predict theincidence and prevalence of persons with ESRD up until theyear 2040.

2. Methodology

This study was registered and approved by the ethics com-mittee of the National Medical Research Register (NMRR-13-1783-18577). Univariate forecasting modeling was con-structed using the new patients (incidence) with a confirmeddiagnosis of ESRD on renal replacement therapy (RRT) inMalaysia by year, from 1993 to 2013. The univariate forecast-ing models were applied since these techniques also generategood forecasts especially when the trend is stable [4]. Fourtypes of time-series analyses methods were assessed, namely,the naivewith trendmethod, the double exponential smooth-ing method, Holt’s method, and autoregressive integratedmoving average (ARIMA) modeling method. Forecastingperformance is evaluated by estimating the RMSE, MAPE,and MAE and the model with the smallest forecast error wasselected to be used in predicting the incidence and prevalenceof ESRD up until 2040.

The incidence and prevalence rates of ESRDpatients werecalculated using population estimates from the Departmentof Statistics, Malaysia. Meanwhile, based on a previous localstudy in 2001, the cost of dialysis treatment for each patientwas estimated to be RM 30,000 (USD 7,500) per patient, peryear [5]. The cost per patient was estimated to be within RM29,092 to RM 33,642 [5]. The estimated future costs did notincorporate the expected changes due to various factors likechanges in foreign exchange rate, consumer price index,changes in the cost of utilities, emoluments, consumables,hardware, pharmaceuticals, and others. The analysis was

conducted usingMicrosoft Excel and R software (R Develop-ment Core Team (2013). R: A language and environmentfor statistical computing. R Foundation for Statistical Com-puting, Vienna, Austria. ISBN 3-900051-07-0, URL https://www.R-project.org.).

3. Result

The yearly incidence and prevalence of dialysis patients(haemodialysis and peritoneal dialysis) inMalaysia have beenon the increasing trend for the past 20 years. Based onthe four types of univariate models measured, the ARIMAmodeling which gives the smallest RMSE, MAPE and MAEwas selected to be used for predicting the number of newdialysis patients from 2014 to 2040. ARIMA (0, 2, 1) was usedfor both projection of incidence of dialysis patients (RMSE =135.50, MAPE = 2.85, and MAE = 87.71) and also prevalenceof dialysis patients (RMSE = 158.79, MAPE = 1.29, andMAE = 117.21) (Table 1).

Based on the model selected, the projected number ofincidence and prevalence dialysis patients in Malaysia willincrease as shown in Figures 1 and 2.The estimated incidenceof new dialysis patients inMalaysia in 2020 is 10,208 cases and19,418 in 2040. Meanwhile, the estimated prevalence is 51,269and 106,249 cases in 2020 and 2040, respectively. With suchprojected prevalence, the estimated costs for the treatment areUSD 384,517,500 and USD 796,867,500 in the years 2020 and2040, respectively (Table 2).

4. Discussion

Based on projections made by the Department of Statistics(DOS), Malaysia is expected to reach aging population statusby the year 2035, at which point 15 per cent of the totalpopulation will be 60 years and older [6]. As the number ofolder adults is projected to increase dramatically over the nexttwenty years, this will pose major challenges to our healthcare systems, as there will be greater healthcare utilization forcomorbid conditions among this population [7].

With the rise in conditions such as obesity, diabetes, andhypertension in the middle age population, it is likely thatthe future prevalence of chronic kidney disease (CKD) willincrease further among the elderly. It is a significant concernwith both an increasing incidence of treated kidney failurewith dialysis and a high prevalence of earlier stages of CKD.CKD is silent and irreversible. If CKD is not detected early ormanaged properly, the conditionwill progress to ESRDwhichwill result in the need for renal replacement therapy witheither dialysis or kidney transplant. Additionally, ESRD alsocan lead to mental illnesses such as anxiety and depressionwhich will affect the patients’ quality of life [8].

Renal replacement therapy (RRT) in Malaysia has shownan exponential growth since 1990 [9]. The rate of incidenceof end-stage renal disease requiring RRT is forecasted to be3.02 (per 10,000 population) by the year 2020 and to increasefurther to 3.89 in 2030 and reaching a rate of 4.68 in 2040.There is a projected 1.5-fold increase in overall number ofpatients on dialysis from 6,985 in 2013 to 10,208 in 2020 anda further 1.5-fold increase to 14,813 in 2030, with a 1.3-fold

https://www.R-project.org

https://www.R-project.org

International Journal of Nephrology 3

Table 1: Comparison of forecast error from number of ESRD patients, 1993–2013.

Method Incidence PrevalenceParameter RMSE MAPE MAE Parameter RMSE MAPE MAE

Naive with trend — 165.12 3.73 116.37 — 195.81 1.81 160.16Double exponentialsmoothing 𝛼 = 0.6577 151.33 5.64 110.87 𝛼 = 0.9999 205.96 2.72 169.47

Holt 𝛼 = 0.9999,𝛽 = 0.3221

135.40 3.24 90.26 𝛼 = 0.9999,𝛽 = 0.9999

199.78 3.09 167.14

ARIMA (0, 2, 1) — 135.50 2.85 87.71 — 158.79 1.29 117.21

1993–2013 2014–2040

4020 3010 500Number of years

0

5000

10000

15000

20000

25000

30000

Num

ber

Figure 1: Trend and forecast values for incidence of ESRD patients(ARIMA [0, 2, 1]).

1993–2013 2014–2040

0

50000

100000

150000

Num

ber

10 20 30 40 500Number of years

Figure 2: Trend and forecast values for prevalence of ESRD patients(ARIMA [0, 2, 1]).

increase to 19,418 in 2040. This is a serious health and eco-nomic burden, as patients with ESRD consume a vastly dis-proportionate amount of financial and human resources.

The cost of CKD to the National Health System in theUnited Kingdom from the year 2009 to 2010 is estimatedto be m1.44 to m1.45 billion, which is about 1.3% of all NHSexpenditure in that year. More than half of this amountwas spent on RRT, which was provided for 2% of the CKDpopulation [10]. In Malaysia, we projected the costs based oncosts estimated by Hooi and colleagues [5]. They estimatedthe costs were within RM 29,092 to RM 33,642 per patientdialysed atMinistry ofHealth facilities. By taking average costof RM 30,000 per patient, the estimated cost incurred to treatall patients with dialysis in the year 2040 is USD 796,867,500

(including new patients at USD 145,635,000) which is morethan 10 times compared to the amount in the year 2000. Sincethe cost was estimated based on year 2001, we postulate thatthe future cost is higher than what we have estimated dueto various factors such as changes in the currency exchangerate, consumer price index, and possibility of the increasedcosts of the current treatment. Hence, the future costs that isestimated by this study can only become a rough projectionto indicate the magnitude and burden of disease.

Careful projections of the demand for dialysis servicesare important to assist healthcare planners in forecasting theneed for equipment, facilities, and personnel [11]. Even so, thesetting up of more dialysis centres is not the way to go, as itcannot keep up with the rising need for RRT. More effectiveprevention, intervention, and early detection programs forrenal diseases should be put into place to increase awarenessamong the public and medical personnel alike.

The health care system should place an emphasis on edu-cation and on creating awareness on CKD among the generalpublic.They should be aware of the comorbid conditions thatlead to CKD and should learn to adapt lifestyle changes tomanage their health properly. Medical professionals shouldbe trained adequately and expanded surveillance and screen-ing systems for chronic diseases made available and easilyaccessible. The Disease Control Priorities Project offers rec-ommendations on these measures, supporting the establish-ment of international centers of excellence for CKD thatcan work in tandem with centers of cardiovascular diseases(CVD) and diabetes [12]. Also, there are collaborationsbetween the International Society of Nephrology and KidneyDisease Improving Global Outcomes with societies such asthe International Diabetes Federation (IDF), WHO, and theWorld Bank to influence global and national health decision-making with regard to the growing burden of CKD andchronic diseases [13, 14].

On a larger scale, there needs to be better alignmentof disease funding with burden of disease, transparency ofprojects, and data availability and better sharing of knowledgewith more concerted activities to address the problem ofchronic kidney disease (CKD) [15]. More specific cost-effec-tive strategies need to be implemented to cope with the risingdemand for RRT. Specific interventions may be used, forexample, the increased use of angiotensin-converting enzymeinhibitors and angiotensin receptor type 2 antagonists toslow the rate of decline in renal function [16]. In diabetics,glycaemic control may be improved with the use of newer

4 International Journal of Nephrology

Table2:Estim

ated

costto

treatpatie

ntsw

ithES

RDin

year

2000

to2040

.

Year

Est.po

p.(in

’000)

Incidence

Prevalence

Num

ber

Rate(per

10,000

pop.)

Est.cost(RM

mil)

Est.cost(U

SDmil)

Num

ber

Rate(per

10,000

pop.)

Est.cost(RM

mil)

Est.cost(U

SDmil)

2000

23,494.90

1,853

0.79

55.59

13.90

6,702

2.85

201.0

650.27

2005

26,045.50

3,167

1.22

95.01

23.75

13,35

65.13

400.68

100.17

2010

28,588.60

5,305

1.86

159.1

539.79

23,709

8.29

711.2

7177.8

22015

a31,18

6.10

7,906

2.54

237.18

59.30

37,52

412.03

1,125.72

281.4

32020

a33,782.40

10,208

3.02

306.24

76.56

51,269

15.18

1,538.07

384.52

2025

a36,022.70

12,511

3.47

375.33

93.83

65,014

18.05

1,950.42

487.6

12030

a38,062.20

14,813

3.89

444.39

111.10

78,759

20.69

2,362.77

590.69

2035

a39,879.30

17,116

4.29

513.48

128.37

92,504

23.20

2,775.12

693.78

2040

a41,503.10

19,418

4.68

582.54

145.64

106,249

25.60

3,187.4

7796.87

Est.=estim

ated;pop.=

popu

latio

n.ae

stimates

numbero

fcases

from

thes

elected

mod

el(A

RIMA[0,2,1]).

Each

ESRD

patient

isestim

ated

toincurc

ostfor

RM30,000

(onestim

ated

rateof

1.00USD

=4.00

MYR

).Source

ofestim

ated

popu

lationfro

mDepartm

ento

fStatistics,Malaysia

.

International Journal of Nephrology 5

antidiabetics agents, thus lowering the rate of renal loss.Furthermore, antismoking initiatives may also help slow theprogression of chronic kidney disease to ESRD [17].

With these efforts in place, we can perhaps better respondto the challenging burden of CKD and the rising need forRRT in the country. As for conclusions, the predicted numberof incidence and prevalence of dialysis patients in Malaysiaseem to be increasing over the years until 2040. The numberof patients commencing dialysis in Malaysia is predicted tobe more than tripled, from 32,026 in 2013 to 106,249 in 2040(including new patients from 6,985 in 2013 to 19,418 in 2040).The pmp of new dialysis patients in Malaysia is estimated toincrease by almost twofold; from 231 pmp in 2013 to 467 pmpin 2040. In view of these trends especially with the trend ofaging population in the country [7], there is a need for moretargeted, unified, and coordinated efforts to bettermanage theissues that may arise from this pattern.

The limitation of study is that the univariate forecastingmodel was used. Therefore, this model did not incorporateother important factors such as age, prevalence of comor-bidities (i.e., diabetesmellitus and hypertension), and lifestylefactors. Most of these data are difficult to be retrieved;hence, applying a univariate forecasting model especially forstable time-series pattern is necessary. The forecasted inci-dence is assumed to be true based on historical data whenthere is no change in the populations’ behavior andno specificinterventions have been made. However, with public aware-ness and lifestyle changes with effective changes in health-care management, it is possible that in the future the trueincidence and prevalence will be lower than what has beenestimated.

Ethical Approval

This study was registered and approved by the ethics commit-tee of the National Medical Research Register.

Competing Interests

Authors declare no conflict of interests with regard to thepublication of this paper.

Acknowledgments

The authors thank the Director General of Ministry ofHealth for the permission to publish this manuscript and theNational Renal Registry, Malaysia, especially Lee Day Guatfor their contribution and data sharing.

References

[1] L. A. Szczech and I. L. Lazar, “Projecting theUnited States ESRDpopulation: issues regarding treatment of patients with ESRD,”Kidney International, Supplement, vol. 66, no. 90, pp. S3–S7,2004.

[2] The National Renal Registry, “22nd Report of the MalaysianDialysis and Transplant Registry 2014,” 2015.

[3] M. Wakasugi, J. J. Kazama, and I. Narita, “Anticipated increasein the number of patients who require dialysis treatment among

the aging population of Japan,”Therapeutic Apheresis and Dial-ysis, vol. 19, no. 3, pp. 201–206, 2015.

[4] M. A. Bujang, T. H. Adnan, P. Supramaniam, A.M. AbdHamid,and J. Haniff, “Prediction number of deaths by occurrence inMalaysia: a comparison between simple linear regressionmodeland holt’s linear trendmodel,” StatisticsMalaysia—Journal of theDepartment of Statistics, Malaysia, vol. 2, pp. 25–37, 2009.

[5] L. S. Hooi, T. O. Lim, A. Goh et al., “Economic evaluation ofcentre haemodialysis and continuous ambulatory peritonealdialysis in Ministry of Health hospitals, Malaysia,” Nephrology,vol. 10, no. 1, pp. 25–32, 2005.

[6] Department of Statistics Malaysia, “Population Quick Info.Department of Statistics,” http://pqi.stats.gov.my/searchBI.php.

[7] M. A. Bujang, A.M. AbdulHamid, N. A. Zolkepali, N.M.Ham-edon, S. S. Mat Lazim, and J. Haniff, “Mortality rates by specificage group and gender inMalaysia: trend of 16 years, 1995–2010,”Journal of Health Informatics in Developing Countries, vol. 6, no.2, pp. 521–529, 2012.

[8] M. A. Bujang, R. Musa, W. J. Liu, T. F. Chew, C. T. S. Lim, andZ. Morad, “Depression, anxiety and stress among patients withdialysis and the association with quality of life,”Asian Journal ofPsychiatry, vol. 18, pp. 49–52, 2015.

[9] L. S. Hooi, H. S. Wong, and Z. Morad, “Prevention of renalfailure: theMalaysian experience,”Kidney International, Supple-ment, vol. 67, no. 94, pp. S70–S74, 2005.

[10] M. Kerr, B. Bray, J.Medcalf, D. J. O’Donoghue, and B.Matthews,“Estimating the financial cost of chronic kidney disease to theNHS in England,” Nephrology Dialysis Transplantation, vol. 27,no. 3, pp. iii73–iii80, 2012.

[11] R. R. Quinn, A. Laupacis, J. E. Hux, R. Moineddin, M. Paterson,and M. J. Oliver, “Forecasting the need for dialysis services inOntario, Canada to 2011,” Healthcare Policy, vol. 4, no. 4, pp.e151–e161, 2009.

[12] D. Jamison, W. Mosley, and A. Measham, Disease ControlPriorities in Developing Countries, World Bank, Washington,DC, USA, 2006.

[13] R. C. Atkins, “The epidemiology of chronic kidney disease,”Kidney International, Supplement, vol. 67, no. 94, pp. S14–S18,2005.

[14] K. Iseki, “Metabolic syndrome and chronic kidney disease: aJapanese perspective on a worldwide problem,” Journal of Neph-rology, vol. 21, no. 3, pp. 305–312, 2008.

[15] R. A. Nugent, S. F. Fathima, A. B. Feigl, and D. Chyung, “Theburden of chronic kidney disease on developing nations: a 21stcentury challenge in global health,” Nephron—Clinical Practice,vol. 118, no. 3, pp. c269–c277, 2011.

[16] American Diabetes Association, “Standards of medical carefor patients with diabetes mellitus, position Statement. Clinicalpractice recommendations (2001),” Diabetes Care, vol. 24, pp.S33–S43, 2001.

[17] R.H. Grimm Jr., K.H. Svendsen, B. Kasiske,W. F. Keane, andM.M.Wahi, “Proteinuria is a risk factor for mortality over 10 yearsof follow-up. MRFIT Research Group. Multiple Risk FactorIntervention Trial,” Kidney International. Supplement, vol. 63,pp. S10–S14, 1997.

http://pqi.stats.gov.my/searchBI.php

Submit your manuscripts athttps://www.hindawi.com

Stem CellsInternational

Hindawi Publishing Corporationhttp://www.hindawi.com Volume 2014


MEDIATORSINFLAMMATION

of


Behavioural Neurology

EndocrinologyInternational Journal of



Disease Markers


BioMed Research International

OncologyJournal of



Oxidative Medicine and Cellular Longevity


PPAR Research

The Scientific World JournalHindawi Publishing Corporation http://www.hindawi.com Volume 2014

Immunology ResearchHindawi Publishing Corporationhttp://www.hindawi.com Volume 2014

Journal of

ObesityJournal of



Computational and Mathematical Methods in Medicine

OphthalmologyJournal of


Diabetes ResearchJournal of



Research and TreatmentAIDS


Gastroenterology Research and Practice


Parkinson’s Disease

Evidence-Based Complementary and Alternative Medicine

Volume 2014Hindawi Publishing Corporationhttp://www.hindawi.com

forecasting the incidence and prevalence of patients with...

Documents